code.delx.au - webdl/blob - sbs.py

   1 import requests_cache
   2 from common import grab_html, grab_json, grab_xml, download_hls, download_mpd, Node, append_to_qs
   3
   4 import json
   5 import sys
   6
   7 BASE = "https://www.sbs.com.au"
   8 FULL_VIDEO_LIST = BASE + "/api/video_search/v2/?m=1&filters={section}{Programs}"
   9 VIDEO_URL = BASE + "/ondemand/video/single/%s"
  10
  11 NS = {
  12     "smil": "http://www.w3.org/2005/SMIL21/Language",
  13 }
  14
  15
  16 class SbsVideoNode(Node):
  17     def __init__(self, title, parent, url):
  18         Node.__init__(self, title, parent)
  19         self.video_id = url.split("/")[-1]
  20         self.can_download = True
  21
  22     def download(self):
  23         with requests_cache.disabled():
  24             doc = grab_html(VIDEO_URL % self.video_id)
  25         player_params = self.get_player_params(doc)
  26         release_url = player_params["releaseUrls"]["html"]
  27
  28         filename = self.title + ".ts"
  29
  30         hls_url = self.get_hls_url(release_url)
  31         if hls_url:
  32             return download_hls(filename, hls_url)
  33         else:
  34             return download_mpd(filename, release_url)
  35
  36     def get_player_params(self, doc):
  37         for script in doc.xpath("//script"):
  38             if not script.text:
  39                 continue
  40             for line in script.text.split("\n"):
  41                 s = "var playerParams = {"
  42                 if s in line:
  43                     p1 = line.find(s) + len(s) - 1
  44                     p2 = line.find("};", p1) + 1
  45                     if p1 >= 0 and p2 > 0:
  46                         return json.loads(line[p1:p2])
  47         raise Exception("Unable to find player params for %s: %s" % (self.video_id, self.title))
  48
  49     def get_hls_url(self, release_url):
  50         with requests_cache.disabled():
  51             doc = grab_xml("https:" + release_url.replace("http:", "").replace("https:", ""))
  52             video = doc.xpath("//smil:video", namespaces=NS)
  53             if not video:
  54                 return
  55             video_url = video[0].attrib["src"]
  56             return video_url
  57
  58 class SbsNavNode(Node):
  59     def create_video_node(self, entry_data):
  60         SbsVideoNode(entry_data["title"], self, entry_data["id"])
  61
  62     def find_existing_child(self, path):
  63         for child in self.children:
  64             if child.title == path:
  65                 return child
  66
  67 class SbsRootNode(SbsNavNode):
  68     def __init__(self, parent):
  69         Node.__init__(self, "SBS", parent)
  70
  71     def fill_children(self):
  72         all_video_entries = self.load_all_video_entries()
  73         category_and_entry_data = self.explode_videos_to_unique_categories(all_video_entries)
  74         for category_path, entry_data in category_and_entry_data:
  75             nav_node = self.create_nav_node(self, category_path)
  76             nav_node.create_video_node(entry_data)
  77
  78     def load_all_video_entries(self):
  79         offset = 1
  80         amount = 50
  81         uniq = set()
  82         while True:
  83             url = append_to_qs(FULL_VIDEO_LIST, {"range": "%s-%s" % (offset, offset+amount-1)})
  84             data = grab_json(url)
  85             if "entries" not in data:
  86                 raise Exception("Missing data in SBS response", data)
  87             entries = data["entries"]
  88             if len(entries) == 0:
  89                 break
  90             for i, entry in enumerate(entries):
  91                 if entry["guid"] not in uniq:
  92                     uniq.add(entry["guid"])
  93                     yield entry
  94             offset += amount
  95             sys.stdout.write(".")
  96             sys.stdout.flush()
  97         print()
  98
  99     def explode_videos_to_unique_categories(self, all_video_entries):
 100         for entry_data in all_video_entries:
 101             for category_data in entry_data["media$categories"]:
 102                 category_path = self.calculate_category_path(
 103                     category_data["media$scheme"],
 104                     category_data["media$name"],
 105                 )
 106                 if category_path:
 107                     yield category_path, entry_data
 108
 109     def calculate_category_path(self, scheme, name):
 110         if not scheme:
 111             return
 112         if scheme == name:
 113             return
 114         name = name.split("/")
 115         if name[0] != scheme:
 116             name.insert(0, scheme)
 117         return name
 118
 119     def create_nav_node(self, parent, category_path):
 120         if not category_path:
 121             return parent
 122
 123         current_path = category_path[0]
 124         current_node = parent.find_existing_child(current_path)
 125         if not current_node:
 126             current_node = SbsNavNode(current_path, parent)
 127         return self.create_nav_node(current_node, category_path[1:])
 128
 129 def fill_nodes(root_node):
 130     SbsRootNode(root_node)